Frame-Semantic Parsing

By Dipanjan Das, Desai Chen, André F. T. Martins, Nathan Schneider, Noah A. Smith. 2014

333 citations recorded by the source corpus.

Frame semantics is a linguistic theory that has been instantiated for English in the FrameNet lexicon. We solve the problem of frame-semantic parsing using a two-stage statistical model that takes lexical targets (i.e., content words and phrases) in their sentential contexts and predicts frame-semantic structures. Given a target in context, the first stage disambiguates it to a semantic frame. This model uses latent variables and semi-supervised learning to improve frame disambiguation for targets unseen at training time. The second stage finds the target's locally expressed semantic arguments. At inference time, a fast exact dual decomposition algorithm collectively predicts all the arguments of a frame at once in order to respect declaratively stated linguistic constraints, resulting in qualitatively better structures than naïve local predictors. Both components are feature-based and discriminatively trained on a small set of annotated frame-semantic parses. On the SemEval 2007 benchmark data set, the approach, along with a heuristic identifier of frame-evoking targets, outperforms the prior state of the art by significant margins. Additionally, we present experiments on the much larger FrameNet 1.5 data set. We have released our frame-semantic parser as open-source software.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

Multi-Granularity Hierarchical Attention Fusion Networks for Reading Comprehension and Question Answering (2018, methodology): “Restricted by the limited volume of the annotated dataset, early studies mainly rely on a pipeline of NLP models to complete this task, such as semantic parsing and linguistic annotation (Das et al., 2014).”
TGIF: A New Dataset and Benchmark on Animated GIF Description (2016, methodology): “We ﬁrst obtain the semantic roles of words in our training examples by applying a semantic parser [31, 7].”
Machine Comprehension with Syntax, Frames, and Semantics (2015, methodology): “Frame semantic parsing (Das et al., 2014) is the problem of extracting frame-speciﬁc predicate-argument structures from sentences, where the frames come from an inventory such as FrameNet (Baker et al., 1998).”
Automatically Solving Number Word Problems by Semantic Parsing and Reasoning (2015, background): “In semantic role labeling and frame-semantic parsing (Gildea & Jurafsky, 2002; Carreras & Marquez, 2004; Marquez et al., 2008; Baker et al., 2007; Das et al., 2014), predicate-argument structures are discovered from text as their shallow semantic representation.”
When is multitask learning effective? Semantic sequence prediction under varying data conditions (2016, positive): “We use the data splits from (Das et al., 2014; Hermann et al., 2014).”
A Joint Sequential and Relational Model for Frame-Semantic Parsing (2017, methodology): “Our baselines include SEMAFOR (Das et al., 2014) 9 , a widely used frame-semantic parser for English, and SEMAFOR ( BEST ), an improved SEMAFOR system that is trained with heterogeneous resources (Kshirsagar et al., 2015).”
Syntax-based Rewriting for Simultaneous Machine Translation (2015, methodology): “To ensure that rewrites leave meaning unchanged, we use the SEMAFOR semantic role labeler (Das et al., 2014) on the original and mod-iﬁed sentence; for each role-labeled token in the reference sentence, we examine its corresponding role in the rewritten sentence and calculate the average accuracy…”
Dual Decomposed Learning with Factorwise Oracle for Structural SVM of Large Output Domain (2016, background): “For instance, many applications in NLP such as Machine Translation [3], Speech Recognition [21], and Semantic Parsing [1] have output domains as large as the size of vocabulary, for which the prediction of even a single sentence takes considerable time.”
Semantic Role Labeling with Neural Network Factors (2015, methodology): “Similar approaches (Das et al., 2014; Hermann et al., 2014) have been applied to frame-semantic parsing us-ing FrameNet conventions (Baker et al., 1998).”
Efficient Inference and Structured Learning for Semantic Role Labeling (2015, methodology): “, 2008; Das et al., 2012), as well as on ensembles for combining the predictions of separate systems using integer linear programming (Surdeanu et al., 2007; Punyakanok et al., 2008).8 Meza-Ruiz and Riedel (2009) further used a Markov Logic Network formulation to incorporate a subset of these constr”
Augmenting FrameNet Via PPDB (2014, methodology): “The task of frame semantic parsing was introduced by Gildea and Jurafsky (2002) and later it matured into a community-wide shared task (Baker et al., 2007), with CMU’s SE-MAFOR system being the current state-of-the-art parser (Das et al., 2013).”
Syntactic Scaffolds for Semantic Structures (2018, methodology): “…for a very tight integration of syntactic and semantic processing (Steedman, 2000; Copestake and Flickinger, 2000), and many systems have used syntactic dependency or phrase-based parsers as preprocessing for semantic analysis (Gildea and Palmer, 2002; Punyakanok et al., 2008; Das et al., 2014).”
Spatial role labeling with convolutional neural networks (2015, background): “Analogous to semantic role labeling [3, 10, 5], the task of spatial role labeling [17] has been deﬁned as that of automatically labeling words and phrases, in a sentence, with a set of spatial roles such as tra-jector, landmark, spatial indicator, distance, direction, etc.”
Cooking with Semantics (2014, methodology): “We then applied a state-of-the art SRL system (Das et al., 2014) to the corpus, using Propbank (Palmer et al., 2005) as our frame repository.”
Event2Mind: Commonsense Inference on Events, Intents, and Reactions (2018, background): “Prior work has sought formal frameworks for inferring roles and other attributes in relation to events (Baker et al., 1998; Das et al., 2014; Schuler et al., 2009; Hartshorne et al., 2013, inter alia ), implicitly connoted by events (Reisinger et al., 2015; White et al., 2016; Greene, 2007; Rashkin…”
Multi-Task Learning of Keyphrase Boundary Classification (2017, methodology): “…Søgaard and Goldberg (2016); (2) frame target annotations from FrameNet 1.5 (corresponding to the target identiﬁcation and classiﬁcation tasks in Das et al. (2014)); (3) hyperlink prediction using the dataset from Spitkovsky et al. (2010), (4) identiﬁcation of multi-word expressions using the…”
Multitask learning for semantic sequence prediction under varying data conditions (2016, neutral): “We use the data splits from (Das et al., 2014; Hermann et al., 2014).”
Semantic Frame Identification with Distributed Word Representations (2014, methodology): “We have presented a simple model that outperforms the prior state of the art on FrameNet-style frame-semantic parsing, and performs at par with one of the previous-best single-parser systems on PropBank SRL. Unlike Das et al. (2014), our model does not rely on heuristics to construct a similarity gr”